Unemployment During the Great Depression

Regarding unemployment during the Great Depression, Andrew Wilson writing at the WSJ recently said:

As late as 1938, after almost a decade of governmental “pump priming,” almost one out of five workers remained unemployed.

Historian Eric Rauchway says this is a lie, a lie spread by conservatives to besmirch the sainted FDR.  Nonsense.  In 1938 the unemployment rate was 19.1%, i.e. almost one out of five workers was unemployed, this is from the official Bureau of Census/Bureau of Labor Statistics data series for the 1930s. You can find the series in Historical Statistics of the United States here (big PDF) or here.  The graph is at right. Rauchway knows this but wants to measure unemployment using an alternative series which shows a lower unemployment rate in 1938 (12.5%).  Nothing wrong with that but there’s no reason to call people who use the official series liars.

So why are there multiple series on unemployment for the 1930s?  The reason is that the current sampling method of estimation was not developed until 1940, thus unemployment rates prior to this time have to be estimated and this leads to some judgment calls.  The primary judgment call is what do about people on work relief.  The official series counts these people as unemployed.

Rauchway thinks that counting people on work-relief as unemployed is a right-wing plot.  If so, it is a right-wing plot that exists to this day because people who are on workfare, the modern version of work relief, are also counted as unemployed.  Now if Rauchway wants to lower all estimates of unemployment, including those under say George W. Bush, then at least that would be even-handed but lowering unemployment rates just under the Presidents you like hardly seems like fair play.

Moreover, it’s quite reasonable to count people on work-relief as unemployed.  Notice that if we counted people on work-relief as employed then eliminating unemployment would be very easy – just require everyone on any kind of unemployment relief to lick stamps.  Of course if we made this change, politicians would immediately conspire to hide as much unemployment as possible behind the fig leaf of workfare/work-relief.

There is a second reason we may not want to count people on work-relief as employed and that is if we are interested in the effect of the New Deal on the private economy.  In other words, did the fiscal stimulus work to restore the economy and get people back to work?  Well, we can’t answer that question using unemployment statistics if we count people on work-relief as employed.  Notice that this was precisely the context of the WSJ quote.

One final thing that one could do is count people on work-relief as neither employed nor unemployed, i.e. not part of the labor force which is what we do for people in the military.  Rauchway has data on this and it shows almost the same thing, nearly one in five unemployed, as the original series.  (In this case, however, Rauchway counts nearly one in five unemployed as a win for the New Deal because the same series also shows higher unemployment earlier in the Great Depression.)

Any way you slice it there is no right-wing plot to raise unemployment rates during the New Deal and a historian should not go around calling people liars just because their judgment offends his wish-conclusions.

Hat tip to Mark Thoma.


Comments for this post are closed